Highway Safety Manual (HSM) Freeway Training Course

Ohio Department of Transportation John R. Kasich, Governor Jerry Wray, Director Freeway and Interchange Crash Prediction ECAT ODOT HSM Freeway and Interchange Training 2018 Course Objectives and Outcomes Objectives: To inform participants about the latest methods of freeway and interchange safety analysis To demonstrate how to apply these tools to the design and alternatives analysis process Participants should be able to: Apply evaluation tool (ECAT) to typical designs Evaluate the safety performance associated with a design Understand the relationship of key geometric elements of freeways and interchanges to safety performance 2 ODOT Safety Training 1

Session 1 Introduction & Background on Freeways and Interchanges 3 Freeway and Interchange Crash Prediction 4 ODOT Safety Training 2

Challenges for ODOT Interchange and freeway corridor projects are the most complex and expensive of all roadway projects 5 Basic interchange forms vary in cost, footprint and operations.. as well as their safety performance 6 ODOT Safety Training 3

We should understand the expected safety performance of a $250 million investment Would you expect these three alternatives to experience the same number of crashes over a 30 year project life? If not, would it be helpful to understand the potential differences when selecting a preferred alternative? 7 Interchange and Corridor Planning and Design Issues Configuration (service and system) Design geometry (e.g., design speed of ramps) Interchange spacing Weaving vs. CD roads vs. ramp braids Design Level of Service (number of lanes and amount of traffic for which design should accommodate) Impact of Design Waiver/Exceptions 8 ODOT Safety Training 4

HSM Chapter on Freeways and Interchanges AASHTO funded research on freeways and interchanges to fill the 1st edition gap NCHRP Project 17-45 ISATe Interchange Safety Analysis Tool Enhanced Provides Methods and Models for predicting Average Annual Crash Frequency for Freeways and Ramps ISATe Tool released for use in 2012 Published as HSM 2014 Supplement ECAT Tool has been updated to add Freeway Analysis (2016/2017) 9 Crash Prediction for Freeway Segments and Interchanges Statistical modeling process similar to other chapters in HSM was used SPF, CMF, C Data from California, Maine and Washington state Google earth used to obtain horizontal and cross sectional data on hundreds of miles of freeways and interchanges 10 ODOT Safety Training 5

HSM Freeway Analysis Provides High Value in Planning and Design Predict crashes before and after reconstruction of a corridor Evaluate effect of adding new interchange Evaluate effect of increasing capacity of an existing corridor through widening Evaluate effect of increasing or decreasing weaving distance Compare performance of CD vs. mainline weaving vs ramp braid solutions Predict and compare the safety performance of interchange configuration alternatives Evaluate and refine preliminary geometry Evaluate and document design exceptions 11 Freeway Crash Prediction Components Freeway Segments Ramps Ramp Terminal Intersections with Crossroad Intersections Fwy Seg Ramps 12 ODOT Safety Training 6

Key Concept: General form of the HSM predictive methods Predicted Crash Frequency SPF x (CMF 1 x CMF 2 x.) x C Safety Performance Function Crash Modification Factors Local Calibration Factor 13 Key Concept: Safety Performance Function Depending on Site Type (Freeway Segment, Ramp Segment or Ramp Terminal) may consider: Rural vs. Urban Number of Lanes Single Vehicle Multiple Vehicle Fatal and Injury Property Damage Only 14 ODOT Safety Training 7

Key Concept: Crash Modification Factors (CMF) Used to calculate change in crash frequency for a specific change in geometry Adapts SPF to non-base conditions One CMF per design element (e.g. lane width) 13 CMFs for freeways (Table 18-13, pp. 18-35) 9 CMFs for ramp segments (Table 19-22, pp. 19-45) 11 CMFs for crossroad ramp terminals (Table 19-23, pp. 19-46) 15 Divide facility into homogenous segments Most time-intensive part of analysis Most important part of analysis Key Concept: Segmentation The more homogenous a section, the more accurate your results! 16 ODOT Safety Training 8

Key Concept: Calibration Model derived from multiple state databases Reporting thresholds vary state-by-state Uses of an calibrated vs. uncalibrated model Calibrated - preferred ODOT has calibration factors for use in Ohio analysis for many conditions included in ECAT Uncalibrated - still a good analysis tool Compare multiple alternatives against one another K/A reporting is fairly consistent state-by-state Compare predicted crash pattern against observed crash pattern Meaningful for work outside Ohio or where calibration factors don t exist (for conditions not covered by HSM models there will be no calibration factors - ASK ODOT!!!) 17 Data Requirements: Freeway Segments Geometric Data Basic Roadway Data Length, number of lanes, area type Alignment horizontal curve Cross Section lane width, inside/outside shoulder widths, rumble strips, median barrier Roadside data clear zone, roadside barrier Ramp Access presence, side, length of speed-change area Traffic Data AADT - Segment, entrance ramp, exit ramp Proportion of AADT during high-volume hours 18 ODOT Safety Training 9

Data Requirements: Ramp Segments Geometric Data Basic Roadway Data Length, number of lanes, Average Speed on Freeway, type of traffic control at ramp terminal, area type Alignment horizontal curve Cross Section lane width, left/right shoulder widths, lane add/drop in segment Roadside data barrier Ramp Access length of speed-change area, length of weaving section Traffic Data AADT - segment 19 Data Requirements: Ramp Terminal Intersections Geometric Data Basic Intersection Data area type, terminal configuration, presence of non-ramp public street Alignment skew, distance to nearest public street intersection on outside crossroad leg, distance to adjacent ramp terminal Traffic Control presence of protected left/right turn control on crossroad Cross Section crossroad median width, number of lanes crossroad and ramp, right-turn channelization on crossroad, left-turn and right-turn lanes/bays on crossroad Access data number of drives, public street approaches Ramp Access presence, side, length of speed-change area Traffic Data AADT ramp, crossroad 20 ODOT Safety Training 10

Freeway Analysis Limitations Site types not addressed Facilities with HOV lanes Ramp metering Frontage roads Speed change lanes at crossroads Geometric elements not addressed Vertical geometry > 10-lane freeway segments > 2-lane ramp segments Differing barrier types (i.e. cable vs. guardrail vs. jersey barrier) Single point diamond intersection configuration Roundabout ramp terminal intersections Diverging Diamond 21 Special Conditions Freeways with Barrier Separated Managed Lanes Managed Lanes (Express, HOT, HOV) are considered part of median Analysis is performed for General Purpose lanes with managed lane entry/exit points treated as entrance or exit ramps Toll Facilities Can be analyzed provided the section is sufficiently distance from toll facility so that the facility does not influence vehicle operations Areas in immediate vicinity of a toll plaza, where widened to accommodate vehicles through a toll plaza, or areas that experience toll-related traffic queues or speed changes cannot be analyzed via HSM 22 ODOT Safety Training 11

Freeway Analysis Considerations Both Directions of Freeway Analyzed May require averaging of values - see comments/notes in tools Situations with significantly different cross section or differing number of lanes by direction requires special analysis approach ASK ODOT!!! 23 ECAT Limits One file can accommodate up to 75 elements of freeway segments, ramp segments, and/or crossroad ramp terminals An elements is an individual homogenous freeway segment, homogenous ramp segment, or ramp terminal intersection 24 ODOT Safety Training 12

UP NEXT Session 2: Overview of ECAT Tool 27 ODOT Safety Training 14

Session 2 Fundamentals of HSM Predictive Methods for Freeways and Application Using ECAT 1 Freeway Predictive Analysis Evaluation Scope Individual site Freeway segment Ramp or C-D road segment Crossroad ramp terminal Freeway facility Adjacent sites forming freeway and interchange 2 ODOT Safety Training 1

Economic Crash Analysis Tool ODOT spreadsheet to complete HSM calculations. ECAT has the ability to: calculate predicted crash frequencies, complete empirical bayes calculations, predict crash frequencies for proposed conditions, conduct alternatives analyses, and complete benefit-cost analysis. http://www.dot.state.oh.us/divisions/planning/programmanagement/highwaysafety/ HSIP/Pages/ECAT.aspx 3 ECAT Where to Find It? TOOL HELP FILES http://www.dot.state.oh.us/divisions/planning/programmanagement/highwaysafety/ HSIP/Pages/ECAT.aspx 4 ODOT Safety Training 2

Excel Spreadsheet Input The ECAT Tool By individual site element Freeway segment Ramp or CD road segment Crossroad ramp terminal intersection Segmentation is key! Output Crashes for entire facility Crashes by component Distribution of crashes 5 Using the Spreadsheet Tool HSM Toolbox: Follow the steps in the Analysis Processing box. Can open toolbox at any time by clicking toolbox icon. 6 ODOT Safety Training 3

Using ECAT Follow the Steps Basic ECAT Analysis Steps: Create a project file Enter general project information and define elements (segments and intersections) Load crash data from CAM tool (black box) Assign crashes to project elements by location ID (red box) Generate analysis tabs for the segments and intersections (blue box) Enter site data for existing and proposed (if applicable) conditions Generate the summary report and benefit cost (yellow box) 7 ECAT Input Freeway Segments Tab 8 ODOT Safety Training 4

ECAT Input Freeway Segments 9 ECAT Input Ramp Segments 10 ODOT Safety Training 5

ECAT Input Ramp Terminal Intersections 11 Reviewing Results Crashes for Entire Facility 12 ODOT Safety Training 6

Reviewing Results Crashes by Segment 13 Reviewing Results Crashes by Type 14 ODOT Safety Training 7

Freeway Segments Data Requirements Safety Performance Function Crash Modification Factors Examples 2/16/2018 15 Data Requirements: Freeway Segments Geometric Data Basic Roadway Data Length, number of lanes, area type Alignment horizontal curve Cross Section lane width, inside/outside shoulder widths, rumble strips, median barrier Roadside data clear zone, roadside barrier Ramp Access presence, side, length of speed-change area Traffic Data AADT - Segment, entrance ramp, exit ramp Proportion of AADT during high-volume hours 16 ODOT Safety Training 8

Freeway Segment Safety Performance Function Characteristics N spf = a x (AADT) b x L Coefficients a and b by Area type Number of lanes Crash type Severity 2/16/2018 17 Crash Modification Factors Freeway Segment CMFs 13 available Most are functions of geometric variables Developed to work with SPF Match base conditions Calibrated to work in combination (without overestimating safety influence) 18 ODOT Safety Training 9

Freeway Segment Crash Modification Factors Horizontal curve Lane width Shoulder width Inside Outside Median width Barrier Median Roadside High volume (congestion) Lane change (ramp related) Shoulder rumble strips Outside clearance Ramp entrance Ramp exit 2/16/2018 19 Freeway Segment Crash Modification Factors Geometric Inputs and CMFs Each geometric element has one or more CMFs associated with it What do you expect the effect of the following elements would be on crashes (safety performance)? Horizontal Curve Lane Width Weaving Length Location of ramp (right-hand vs. left-hand) CMF > 1.00 means increased predicted crash frequency CMF < 1.00 means decreased predicted crash frequency Table 18-33, pp. 18-35 20 ODOT Safety Training 10

Barrier Data Median Barrier Continuous Offset from traveled way Pieces Length and offset 21 Median Width Cable Median Barrier Width = 1 Foot 22 ODOT Safety Training 11

Median Barrier Example 0.02 mi 0.06 mi 5 ft 14 ft 14 ft 5 ft 0.05 mi 0.03 mi 23 Median Barrier Example Four pieces 1. Length (L ib,1 ) = 0.05 mi, Offset (W off,in,1 ) = 14 ft 2. Length (L ib,2 ) = 0.03 mi, Offset (W off,in,2 ) = 5 ft 3. Length (L ib,3 ) = 0.02 mi, Offset (W off,in,3 ) = 5 ft 4. Length (L ib,4 ) = 0.06 mi, Offset (W off,in,4 ) = 14 ft Add pieces with the same offset 24 ODOT Safety Training 12

Median Barrier Example Add length of pieces with the same offset Four pieces becomes two pieces 1. Length (L ib,1+4 ) = 1.10 mi, Offset (W off,in,1+4 ) = 14 ft 2. Length (L ib,2+3 ) = 0.05 mi, Offset (W off,in,2+3 ) = 5 ft 25 High Volume Volume Variation During Average Day Proportion of AADT during hours where the volume exceeds 1,000 veh/h/ln Proportion > 0.0 if: Hourly volumes continuously high, or A few very high peak hours Use nearest traffic count station data 26 ODOT Safety Training 13

Base Condition Proportion = 0.0 Limits 0.0 to 1.0 Notes Default values available in ISATe High Volume 27 Lane Change Weaving Section Types Type B Other Types 2/16/2018 28 ODOT Safety Training 14

Lane Change Ramp Location and AADT Need for upstream entrance and downstream exits (Not upstream exits and downstream entrances) 29 Ramp Location within Segment Ramp Location Ramp in segment has x = 0.0 Still need AADT 30 ODOT Safety Training 15

Interchange Ramps Data Requirements Safety Performance Function Crash Modification Factors Examples 31 Data Requirements: Ramp Segments Geometric Data Basic Roadway Data Length, number of lanes, Average Speed on Freeway, type of traffic control at ramp terminal, area type Alignment horizontal curve Cross Section lane width, left/right shoulder widths, lane add/drop in segment Roadside data barrier Ramp Access length of speed-change area, length of weaving section Traffic Data AADT - segment 32 ODOT Safety Training 16

Interchange Ramps Segment-Based Evaluation Method works for all configurations 33 Interchange Ramps Safety Performance Function Characteristics N spf = a x (AADT)b x L Coefficients a and b by Exit, entrance, C-D road Area type Number of lanes Crash type Severity 2/16/2018 34 ODOT Safety Training 17

Interchange Ramps Crash Modification Factors Ramp Segment CMFs 9 Available Most are functions of geometric variables Developed to work with SPF Match base conditions Calibrated to work in combination (without overestimated safety influence) Table 19-22, pp. 19-45 35 Interchange Ramps Crash Modification Factors Ramp or C-D Road Horizontal curve Lane width Shoulder width Right/Left Barrier Right/Left Ramp speed-change lane Lane add or drop C-D Road Weaving section 2/16/2018 36 ODOT Safety Training 18

Horizontal Curves (Ramps) Curve Location Curve Speed Prediction Used to estimate curve entry speed Based on curves and tangents encountered Milepost 0.0 location 37 Horizontal Curves (Ramps) Curve Location Milepost = distance along ramp to start of curve Establish one milepost 0.0 for all segments on ramp Establish one milepost 0.0 for all ramps with a common ramp entrance or exit location 38 ODOT Safety Training 19

Horizontal Curves (Ramps) Curve Location Special Case Two milepost 0.0 for some segments Goal is to estimate average curve entry speed on these segments For these segments, use milepost 0.0 of the higher volume ramp 39 Crossroad Ramp Terminal Intersections Intersection Configurations Data Requirements Safety Performance Functions Crash Modification Factors Examples 40 ODOT Safety Training 20

Ramp Terminal Intersection Configurations Seven configurations D3ex: Diagonal Exit D3en: Diagonal Entrance D4: Diagonal 4-Leg 41 Data Requirements: Ramp Terminal Intersections Geometric Data Basic Intersection Data area type, terminal configuration, presence of non-ramp public street Alignment skew, distance to nearest public street intersection on outside crossroad leg, distance to adjacent ramp terminal Traffic Control presence of protected left/right turn control on crossroad Cross Section crossroad median width, number of lanes crossroad and ramp, right-turn channelization on crossroad, left-turn and right-turn lanes/bays on crossroad Access data number of drives, public street approaches Ramp Access presence, side, length of speed-change area Traffic Data AADT ramp, crossroad 42 ODOT Safety Training 21

Ramp Terminal Intersection Configurations Cont. A4: 4-Quad Parclo A B4: 4-Quad Parclo B A2: 2-Quad Parclo A B2: 2-Quad Parclo B 43 Ramp Terminal Intersection Safety Performance Functions Characteristics N spf = a x (AADT xrd ) b x (AADT ex + AADT en ) d AADT xrd = crossroad traffic AADT ex = entrance ramp traffic AADT en = exit ramp traffic Coefficients a and b by Configuration Type of control (signal, stop) Area type Number of crossroad lanes Severity 2/16/2018 44 ODOT Safety Training 22

Ramp Terminal Intersction Safety Performance Function D4 Configuration One-way stop Signal 45 Crash Modification Factors Crossroad Ramp Terminal CMFs 11 Available Most are functions of geometric design or traffic control variables Developed to work with SPF Table 19-23, pp. 19-46 46 ODOT Safety Training 23

Ramp Terminal Intersection Crash Modification Factors Signal or Stop Exit ramp capacity Crossroad turn lane Left turn Right turn Access point frequency Segment length Median width Signal Protected-only leftturn phase Channelized right-turn Crossroad Exit ramp Non-ramp leg Stop Skew Angle 2/16/2018 47 HSM Freeway Analysis Project Example I-270/US 33 Improvements: Dublin, OH Interchange is unique Operates as a service interchange to the east as it approaches the Frantz Road/Post Road intersection Operates as system interchange to the west FREEWAY SYSTEM LOCAL ROADWAY 48 ODOT Safety Training 24

Project Application I-270/US 33 Interchange Goals Improve Safety Address Traffic Congestion Resolve Obsolete Geometric Designs Fiscal responsibility - Develop phased plan to meet funding constraints Three alternatives further studied and developed Alternative 4 Alternative 7 Alternative 8 49 I-270/US 33 Improvements: Alternative 4 50 ODOT Safety Training 25

I-270/US 33 Improvements: Alternative 7 Parclo A interchange 51 I-270/US 33 Improvements: Alternative 8 52 ODOT Safety Training 26

Crash Analysis, 2015-2035 Crash Predictions HSM tool for the I-270/US 33 Interchange The model was uncalibrated as used The results used for comparisons are relative Focused on KAB type crashes from 2015-2035 Most important crash types Reliability of data is greater Safety was one of many criteria used to determine the preferred alternative 53 Results Total KAB predicted crashes 2015-2035 Existing: 308 crashes Alt. 4: 323 crashes Alt. 7: 360 crashes Alt. 8: 320 crashes Societal costs 2015-2035 Existing: $97M Alt. 4: $90M Alt. 7: $102M Alt. 8: $88M 54 ODOT Safety Training 27

Interpretation Trade-off of reconfiguring interchanges with high speed ramp designs Increases VMT Alternative 2015 VMT 2035 VMT Existing 646,908 849,555 Alt 4 910,494 1,198,120 Alt 7 907,212 1,194,732 Alt 8 906,375 1,193,947 Higher quality design Safety performance is better even though VMT is 30% greater than existing 55 Alternative 8 Preferred Alternative(Phase 2) REMOVE LOOP RAMP PHASE 2 CONSTRUCTION (in 10-15 years) 56 ODOT Safety Training 28

Loop Ramp Crash Prediction ISATe used to predict ramp KAB type crashes from 2015 to 2025 Option 1 Maintain existing ramp with a 230 radius Option 2 Reconstruct ramp with a 200 radius Option 3 Reconstruct ramp with a 185 radius 57 Loop Ramp Crash Prediction Per the HSM, comparing the predicted crashes is the appropriate approach for the analysis Using an uncalibrated model is an accepted analysis method From HSM work in Ohio, the KAB crashes are very close to model predictions 58 ODOT Safety Training 29

Summary Model predicts that the KAB crashes for the options are anywhere from 9-10 crashes on the ramp proper over a 10 year period Little difference between the options as far as crash performance Because weaving movements are removed, the main issue will be the speed entering the ramp curve Speed will have a significant influence on safety performance regardless of option 59 Summary Existing ramp radius will be maintained for the project Treatments will be implemented to slow traffic on the approach These countermeasures (or lack thereof) would be expected to have a more appreciable influence on expected crash performance than the ramp radius 60 ODOT Safety Training 30

Questions and Discussion 61 UP NEXT Session 3: ECAT Demonstration 62 ODOT Safety Training 31

Session 3 ECAT Demonstrations Example 1: Urban Mainline and Weaving Section (2en, 2ex) Example 2: Urban Mainline and Weaving Section (2en, 1ex) 1 Session 3 Overview Terminology Getting started with ECAT Overview of the Input Freeway Segments tab Example analyses Discussion of results 2 ODOT Safety Training 1

Session 3 Terminology Terminology Begin/End Milepost Increasing Milepost Decreasing Milepost Project Limits S-C ( speed-change ) Lane 3 Enable macros Main tab Setting Up ECAT 4 ODOT Safety Training 2

Workflow Step 1: Identify analysis limits Step 2: Define study period Consecutive years for which an estimate of the expected average crash frequency is desired Step 3: Acquire data Step 4: Segment project into individual sites Freeway segment Ramp or CD segment Crossroad ramp terminal Step 5: Enter data for each element (segments and intersections) Step 6: Run analysis 5 Freeway Segmentation A new segment begins where there is a change in: Number of Through lanes Lane Width Outside Shoulder Width Inside Shoulder Width Median Width Ramp Presence (specifically the speed-change area) Clear Zone Width 6 ODOT Safety Training 3

Example Freeway Segmentation Homogenous geometry No more than 1 entrance and 1 exit in each segment (in each direction) If comparing multiple alternatives, make sure that the start and end points of each alternative are identical. Segment limits between these points may vary. 7 Segmentation and Through Lanes A new segment should be started at the gore point of a ramp if a lane is added or dropped at the ramp or CD road The number of through lanes for a segment for a ramp add or drop area defined as follows: For the freeway segment use the total number of through lanes in both directions Do not include auxiliary lanes in a weaving section unless the weaving section exceeds 0.85 miles (4.500 ft) Do not include a parallel acceleration lane associated with a ramp that merges with or diverges from the freeway unless it exceeds 0.30 miles (1,600 feet) 8 ODOT Safety Training 4

Freeway Through Lanes at Ramps Begin segment at upstream start of taper Median Segment Segment length, L L add,seg L drop,seg Segment length, L Lane drop Median Lane add Segment Number of through lanes: 5 (= downstream lane count) End segment at upstream start of taper Number of through lanes: 4 (= downstream lane count) For a speed change lane use the number of through lanes in the portion of the freeway adjacent to the speed change lane plus freeway lanes in the opposing travel direction A segment with a lane add or lane-drop taper should be entered with the same number of through lanes as the freeway just downstream of the lane-add or lane-drop 9 Freeway Through Lanes at Ramps PLAN VIEW Ramp Entrance Length, L en Ramp Exit Length, L ex SCen SCex Fr1 Fr2 Fr3 L fs1 L fs2 L fs3 COMPONENT PARTS Speed-Change Lane Speed-Change Lane Type: ramp entrance SCen Type: ramp exit Seg. length = L en Seg. length = L ex SCex Freeway Segment Effective segment length, L * = L fs - L en/2 - L ex/2 (note: freeway segment length does not include the length of speed-change lanes, if these lanes are adjacent to the segment) Fr1 Fr2 Fr3 Otherwise, begin segment at the gore point if the lane is added or dropped at a ramp or C-D road 10 ODOT Safety Training 5

Ramp Spacing and Taper Lengths Ramp in segment Define type (lane add or speed change lane) Define length of entrance from gore tip to 2 separation Define side (left or right) Ramp outside of segment Entrance: Define distance from begin milepost of the current segment to 2 wide gore point of previous upstream entrance ramp Exit: Define distance from end milepost of the current segment to 2 wide gore point of next downstream exit ramp 11 Geometric Input Tolerances Detailed in HSM Cut a new segment if: Number of lanes changes Lane width changes by 0.5 feet or more Shoulder width changes by 1 foot or more Ramp is present Clear zone width changes by 5 feet or more Median width starts changing, stops changing, first exceeds 90 feet, first drops below 90 feet 12 ODOT Safety Training 6

Handout Exercise #1: Urban Freeway Segment (2en, 2ex) N 13 Exercise #1 Summary Results 14 ODOT Safety Training 7

Handout Exercise #2: Urban Freeway Segment (2en, 1ex) N Same freeway Removed increasing milepost exit ramp Shortened weaving section between decreasing milepost ramps Note same overall analysis boundaries What will be the predicted change in safety performance? 15 Exercise #2 Summary Results 16 ODOT Safety Training 8

Comparison & Discussion 17 Questions and Discussion 18 ODOT Safety Training 9

UP NEXT Session 3 cont.: More ECAT Demonstrations 19 ODOT Safety Training 10

Session 3 cont. More ECAT Demonstrations Example 3: Ramp Segment Example 4: Ramp Terminal Configuration Comparison 1 Ramp Configurations Segment-based evaluation Works for all ramp configurations up to 2 lanes Similar to freeway segment analysis Can also analyze C-D roads 2 ODOT Safety Training 1

Mileposting Used to model the speed of a vehicle along the ramp Establishes ramp horizontal curve CMF Milepost = distance along ramp to start of curve Establish one milepost 0.0 for all segments on ramp Establish one milepost 0.0 for all ramps with a common entrance or exit location 3 Mileposting: Special Case Some ramps have two possible milepost 0.0 locations Goal is to estimate average curve entry speed on these segments Use milepost 0.0 of the higher volume ramp 4 ODOT Safety Training 2

Geometric Input Tolerances Cut a new segment if: Number of lanes changes Lane width changes by 0.5 feet or more Shoulder width changes by 1 foot or more Merging/diverging ramp is present 5 Exercise #3: Freeway Ramp Segments 6 ODOT Safety Training 3

Exercise #3 Summary Results 7 Ramp Terminal Analysis Most rigid part of ECAT analysis Method addresses seven different configurations from drop-down menu 8 ODOT Safety Training 4

Handout Exercise #4: Ramp Terminal Configuration Comparison From Freeway ADT = 5000 ADT = 2500 From Freeway ADT = 5000 To Freeway ADT = 7500 ADT = 7500 FREEWAY ADT = 7500 ADT = 7500 FREEWAY To Freeway To Freeway ADT = 2500 A4 Configuration ADT = 5000 D4 Configuration 9 Exercise #4 Characteristics A4 Configuration Signal control ¼ mile to next intersection 800 feet between terminals Crossroad 12 median width 4 lanes both approaches 2 lane exit ramp approach Right turns yield No right turn bays D4 Configuration Signal control ¼ mile to next intersection 800 feet between terminals Left turns protected Channelized 12 foot lanes Right turns yield Channelized Crossroad 12 median width 4 lanes both approaches 2 lane exit ramp approach Exit ramp approach channelized No outside approach right turn bay 1 driveway on outside crossroad 10 ODOT Safety Training 5

Comparison & Discussion A4 Ramp Terminal KA B C O Total Npredicted - Existing Conditions 0.0658 0.4738 0.7763 2.7179 4.0338 11 Comparison & Discussion D4 Ramp Terminal KA B C O Total Npredicted - Existing Conditions 0.0710 0.2930 0.4528 3.9634 4.7802 12 ODOT Safety Training 6

Questions and Discussion 13 UP NEXT Session 4: Full Interchange Example 14 ODOT Safety Training 7

Session 4 Full Interchange Example Example 5: Full Interchange 1 Putting It All Together ECAT can analyze combinations of ramps, ramp terminals, and freeway segments Use in the planning phase when comparing various alternatives What happens when you switch interchange configurations (diamond to a parclo?) What happens when you change the spacing between proposed interchanges? What about adding/removing access to a location entirely? An uncalibrated model is still useful for alternative comparison! 2 ODOT Safety Training 1

Analyzing Output Output Summary tab Breaks down crash types and severity for entire facility by facility component (freeway, ramps, and crossroad terminals) Other output tabs Shows segment-by-segment CMF development and crash number and type development for freeway segments, ramp segments, and ramp terminals CMF of 1.00 is the base condition CMF > 1.00 means increased predicted crash frequency CMF < 1.00 means decreased predicted crash frequency Can use these tabs to isolate factors that directly impact the safety of a specific segment Also a good place to QC: look for unusually high/low CMFs and ask yourself: does this make sense? 3 Best Practices Analysis is only as good as your inputs Record any and all assumptions on a separate spreadsheet as a QC tool Homogenous segmentation is key to a good analysis Allows you to properly identify factors leading to safety issues Produce a segmentation diagram during the analysis process Can be quickly derived from a single-line drawing in the planning phase Allows for easy visual comparison between spreadsheet and actual geometric information Can act as an exhibit in design reports Refer to the manual and previous slides for a list of advice about cutting segments Sometimes you cannot cut a perfect segment again, record any and all assumptions 4 ODOT Safety Training 2

Example Assumptions Sheet 5 Segmentation Diagram Best Practices Indicate type of segment (ramp, ramp taper in a freeway segment, freeway segment, ramp terminal) by color Number each segment to correspond to ECAT worksheet Indicate traffic volumes by segment Indicate direction of increasing/decreasing milepost and direction of traffic flow Different engineers may have different workflows to segment a facility Segment entire facility prior to starting analysis Segment facility as analysis is conducted Benefits/drawbacks to each? 6 ODOT Safety Training 3

Example Segmentation Diagram 7 Pre-Reconstruction: University Street Interchange, Peoria, IL 8 ODOT Safety Training 4

Workshop Exercise Segment and analyze the interchange given the data shown in your worksheet packet As you work, think about how best to leave a record of your assumptions and method of analysis for future QA/QC 9 Pre-Reconstruction Condition Segmentation 10 ODOT Safety Training 5

Pre-Reconstruction Condition Output KA B C O Total Npredicted - Existing Conditions 1.1664 4.3819 4.6693 32.7776 42.9952 11 Post-Reconstruction Condition Segmentation 12 ODOT Safety Training 6

Post-Reconstruction Predictions What do you expect the postreconstruction prediction will be? Total crashes Crashes by severity Freeway vs. ramp crashes 13 Post-Reconstruction Condition Output KA B C O Total Npredicted - Existing Conditions 1.1664 4.3819 4.6693 32.7776 42.9952 Npredicted Proposed Conditions 1.1102 4.2922 4.7580 23.0896 33.2500 14 ODOT Safety Training 7

Pre-construction versus Post-construction total crashes by segment type 15 Output Comparison Reduction in mainline crashes predicted in post-construction Similar ramp crashes predicted in both scenarios More multi-vehicle crashes predicted in prereconstruction Single vehicle crashes remain almost the same Reduction in minor injury and PDO crashes in post-construction 16 ODOT Safety Training 8

Predicted Crash Comparisons: Freeway Segment 2 Pre-Construction Post-Construction WB Weave WB Weave Removed 17 Predicted Crash Comparisons: Freeway Segment 2 2/16/2018 18 ODOT Safety Training 9

Predicted Crash Comparisons: Ramp 4 Pre-Construction Post-Construction One Lane Ramp 2 nd exit removed Two Lane Ramp 19 Predicted Crash Comparisons: Ramp 4 2/16/2018 20 ODOT Safety Training 10

Discussion Why would we see a decrease in predicted crashes? 21 Final Questions and Discussion 22 ODOT Safety Training 11

ISATe Example Exercises Exercise 1: Mainline Freeway Segment 1 (Session 3) Exercise 2: Mainline Freeway Segment 2 (Session 3) Exercise 3: Ramp Segment (Session 3) Exercise 4: Ramp Terminal Configuration (Session 3) Exercise 5: Full Interchange (Session 4)

Session 3 Exercise 1: Mainline Freeway Segment 1 2 P age

3 P age

Session 3 Exercise 2: Mainline Freeway Segment 2 4 P age

Session 3 Exercise 3: Ramp Segment 5 P age

TYPICAL SECTIONS 6 P age

Session 3 Exercise 4: Ramp Terminal Configuration A4 Configuration Signal control ¼ mile to next intersection 800 feet between terminals Crossroad o 12 median width o 4 lanes both approaches o 2 lane exit ramp approach Right turns yield No right turn bays D4 Configuration Signal control ¼ mile to next intersection 800 feet between terminals Left turns protected o Channelized o 12 foot lanes Right turns yield o Channelized Crossroad o 12 median width o 4 lanes both approaches o 2 lane exit ramp approach Exit ramp approach channelized No outisde approach right turn bay 1 driveway on outside crossroad 7 P age

A4 Configuration D4 Configuration 8 P age

Session 4 Exercise 5: Full Interchange PRE RECONSTRUCTION CONFIGURATION ASSUMPTIONS Design Element Dimensions/ Assumptions Mainline number of basic lanes 4 Mainline lane width 12 Mainline outside shoulder width 10 Mainline inside shoulder width 4 Mainline median width 72 Clear zone 30 Mainline median barrier width 2 or 56' (segment 3) Nearest distance from edge of travelled way to median barrier face 8 Outside barrier offset from edge of shoulder 2 Rumble strips Not present Ramp right shoulder width 9 Ramp left shoulder width 4 Diamond ramp lane width 16 Loop ramp lane width 17.5 Average freeway traffic speed 60 MPH POST RECONSTRUCTION CONFIGURATION ASSUMPTIONS Design Element Dimensions/ Assumptions Mainline number of basic lanes 6 west end and core, 8 east end Mainline lane width 12 Mainline outside shoulder width 10 Mainline inside shoulder width 10 Mainline median width Varies, 40 52 (40', 48', 52') Clear zone 30 Mainline median cable barrier width 1 Nearest distance from edge of travelled way to median barrier face 15 Nearest distance from edge of travelled way to outside shoulder barrier face 12 Outside barrier offset from edge of shoulder 2 Rumble strips Not present Ramp right shoulder width 6 Ramp left shoulder width 4 Diamond ramp lane width 16 2 lane ramp lane width 12 2 lane ramp right shoulder width 8 2 lane ramp left shoulder width 6 Loop ramp lane width 16 Average freeway traffic speed 60 MPH

R R R R C R R R G R G SIGN C G G G R G G G G R G G R R R R R R R R R R R G G G G POT Sta 209+67.30 PT Sta 207+95.11 R R G R R R R S G G C S R NEBRASKA AVE PP-118 SIGN PP-116 PT Sta 308+73.95 607+98.55 POT Sta 608+50.40 PT Sta PEORIA CITY OF R NEBRASKA AVE R R R R R C G C G G G G G R G G G S G R R R R R S S 410 15 C PCC 10 PC PC 10 FAI 74 C 15 FAI 74 BILLBOARDS FACILITY G G SIGN 20 20 1 RAMP NUMBER 1 2 AADT 325 3200 3 1950 4 3150 5 3150 6 3250 7 2650 0 200 400 SCALE IN FEET 2347R 2303R DI BLEACHERS BLEACHERS PEORIA MAINLINE 42200 AADT CONC OBSTRUCTIONS WE 527.97 POT Sta 200+00.00 PIPE CONC 200 25 110 PT Sta 110+98.97 461R PC Sta 200+87.19 25 PT Sta 204+09.70 205 450R RAMP 1 PC Sta 106+11.88 PT Sta 105+07.29 221R PC Sta 205+32.57 RAMP 2 605 RAMP 30 RA M P PC Sta 103+74.83 CEMETERY R R R OVERHEAD SIGN SIGN FAI 74 OVERHEAD SIGN FAI 74 UNIVERSITY AVE DRY RUN CREEK FAI 74 PEORIA G R OVERHEAD SIGN PEORIA DRY RUN CREEK R R G R R R G G G G G G UNIVERSITY AVE G PEORIA 300 PC Sta 300+71.63 POT Sta 300+00.00 PCC Sta 301+89.86 PCC Sta 601+46.71 PCC Sta 304+25.85 RAMP 600 3 BROADWAY ST FAI 74 CEMETERY BIT TS NEBRASKA CONT AVE BOX OVERHEAD SIGN OVERHEAD SIGN 305 6 400 500 7 700 RAMP 705 RAMP 4 PT Sta 403+25.78 PCC Sta 401+09.29 PC Sta 400+00.00 405 105 415R 505 PC Sta 405+72.30 PT Sta 406+83.84 PC Sta 100+43.61 PC Sta 408+22.51 POT Sta 100+00.00 100 POT Sta 410+32.82 PT Sta 409+94.08 UG UTIL UNIVERSITY AVE 50 55 60 65 70 UNIVERSITY AVE PCC PCC POT Sta 50+00.00 POT Sta 73+97.16 231R 159R 221R PC Sta 600+00.00 PCC Sta 605+65.24 546R 220R 1238R 35 40 45 5 30 350R PT Sta 33+09.58 2 3 35 40 45 POT Sta 707+89.40 PT Sta PCC Sta 500+47.86 PC Sta 500+00.00 PT Sta 32+43.58 707+44.96 POT Sta 49+30.06 232R PT Sta 507+18.60 PCC Sta 705+50.25 302R 952R POT Sta 48+74.22 226R PCC Sta 505+72.46 PC Sta 700+00.00 150R PCC Sta 701+69.73 167R 150R PT Sta 102+07.17 216R 139R 144R N Segment 1 Sta 15+00 to Sta 27+00 Segment 2 Sta 27+00 to Sta 39+00 Segment 3 Sta 39+00 to Sta 54+00 INCREASING MILEPOST

POT Sta 216+61.98 PT Sta 214+34.06 215 x PT Sta 309+47.98 POT Sta 317+00.71 POT Sta 615+09.69 PT Sta 755R 611+ 16.38 315 310 615 755R 15 15 10 10 PC Sta 10+00.00 PC Sta 10+00.00 PT Sta 203+65.34 2396R PC Sta 200+00.00 20 200 20 1 RAMP NUMBER 1 2 AADT 325 3200 3 1950 4 6300 6 3250 7 2650 0 200 400 SCALE IN FEET 115 2254R PCC Sta 114+09.13 PT Sta 117+20.08 MAINLINE 42200 AADT 755R PC Sta 206+26.10 205 25 25 PT Sta 209+60.01 1148R 110 PC Sta 211+70.66 RAMP 210 2 PCC Sta 107+62.46 279R 164R RAMP 6 30 591R PCC Sta 105+74.21 605 328R PCC Sta 603+42.93 PCC Sta 601+95.29 656R PCC Sta 609+55.94 PCC Sta 601+25.06 610 600 30 PT Sta PT Sta 33+07.51 32+47.98 PT Sta 713+53.28 PC Sta 712+26.31 RAMP 105 1 410R PCC Sta 101+88.73 753R PC Sta 100+00.00 100 POT Sta 413+10.08 RAMP 3 7 RAMP RAMP 4 UNIVERSITY AVE N PC Sta 300+00.00 INCREASING MILEPOST 300 917R PCC Sta 301+95.28 459R PCC Sta 305+96.82 755R 305 50 35 40 45 35 40 45 PC Sta 47+25.18 PC Sta 48+41.82 50 700 2 3 Segment 1 Sta 15+00 to Sta 27+00 Segment 2 Sta 27+00 to Sta 39+00 Segment 3 Sta 39+00 to Sta 54+00 710 400 PC Sta 700+00.00 405 656R PT Sta 710+59.42 PCC Sta 709+04.58 701+46.07 PCC Sta 404+71.42 755R PCC Sta 328R PC Sta 402+73.05 POT Sta 400+00.00 702+62.38 PT Sta 407+00.73 410R PCC Sta 164R 705 410 PC Sta 408+80.97 PT Sta 411+22.22 291R